Diagnostic performance of ¹⁸F-prostate-specific membrane antigen-1007 PET/computed tomography and multiparametric MRI in prostate cancer: application of the PRIMARY score in a sub-Saharan African cohort

Background – Multiparametric MRI (mpMRI) using the Prostate Imaging Reporting and Data System (PI-RADS v2.1) is widely used for prostate cancer evaluation but is limited by interobserver variability. The PRIMARY score standardizes intraprostatic prostate-specific membrane antigen (PSMA) PET interpretation; however, data using 18F-PSMA-1007, particularly in sub-Saharan Africa, remain limited. Purpose – To compare the diagnostic performance of 18F-PSMA-1007 PET/computed tomography (CT) and mpMRI and evaluate the association of primary lesion standardized uptake value (SUVmax) and PRIMARY score with adverse clinicopathologic features. Methods – This retrospective study included 61 men with biopsy-confirmed prostate cancer who underwent both 18F-PSMA-1007 PET/CT and mpMRI between January 2023 and February 2025. PRIMARY and PI-RADS scores were assigned from PET/CT and mpMRI, respectively. Histopathology served as the reference standard. Sensitivity, positive predictive value (PPV), diagnostic accuracy, and associations between SUVmax and clinicopathologic features were analyzed. Results – PET/CT demonstrated sensitivity of 88.3% and diagnostic accuracy of 86.9%, comparable to mpMRI (86.7 and 85.3%, respectively). Both modalities showed a PPV of 98.1%. SUVmax greater than or equal to 12 was significantly associated with nodal metastases (P = 0.040), distant metastases (P = 0.028), higher Gleason score (P = 0.015), elevated prostate specific antigen (PSA; P = 0.018), seminal vesicle invasion (P = 0.001), and bladder or rectal invasion (P = 0.036). Higher PRIMARY scores also trended toward higher Gleason grade and PSA. Conclusion – 18F-PSMA-1007 PET/CT demonstrated diagnostic performance comparable to mpMRI. SUVmax greater than or equal to 12 (PRIMARY score 5) was associated with aggressive clinicopathologic features, supporting the applicability of the PRIMARY score framework in this sub-Saharan African cohort.